An intelligent transport capacity scheduling method and system

Abstract

The application discloses an intelligent transport capacity scheduling method and system, and relates to the technical field of transport capacity scheduling.The intelligent transport capacity scheduling system comprises a transport capacity information integration module and a transport capacity scheduling module.The application calculates the priority index of each order to be distributed, and orders all the orders according to the priority, wherein the priority ordering not only considers the urgency of the order, but also comprehensively evaluates factors such as the goods attribute, the cabin capacity and the green transportation, and forms a comprehensive priority index through weighted calculation, thereby improving the accuracy of the scheduling decision; through the group game rule, the transport party optimizes the bidding strategy according to the priority of the order and the scheduling demand, enhances the flexibility and diversity of the scheduling, and adjusts the bidding weight according to different strategies in the game process, so that all parties can dynamically adjust the scheduling scheme under the condition of limited resources.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of transport capacity scheduling, and particularly relates to an intelligent transport capacity scheduling method and system. BACKGROUND

[0002] Transport capacity scheduling optimizes resource allocation and task allocation to improve transportation efficiency, reduce costs, and ensure timeliness and reliability of the logistics process. Existing transport capacity scheduling technical solutions mainly rely on real-time acquisition of transportation task demand information and existing transport capacity resources, and through data analysis and intelligent algorithms, resource optimization matching is performed. These solutions fail to fully consider the different needs of various parties in the maritime freight scenario, such as the needs of ports for berths, the requirements of carriers for routes and ship configurations, the concerns of shippers for timeliness and safety, and the monitoring and coordination of the system in the scheduling process. These factors often cannot be effectively integrated and optimized in existing solutions.

[0003] Therefore, there is a need to optimize transport capacity resource matching methods, not only considering the initiative, individualized needs and environmental protection requirements of transporters, but also improving overall transportation efficiency, crew satisfaction and the adaptability of the transport capacity scheduling system. SUMMARY

[0004] The present application aims to provide an intelligent transport capacity scheduling method and system that takes into account the needs of various parties in the transport capacity scenario.

[0005] An intelligent transport capacity scheduling method, comprising the following steps:

[0006] In the current transport capacity scheduling scenario, there are a number of orders to be allocated, transporters and transport ports;

[0007] Obtain a set of orders to be allocated, extract the cargo attribute data, route path data and corresponding storage time limit of the transport port to which each order to be allocated belongs. Obtain a set of all cargo ships of the transporters, extract the real-time cabin capacity, cargo ship draft depth and green transportation index of each cargo ship. Obtain a set of transport ports, extract the real-time water level data and water level prediction values in the next N time slices of each transport port;

[0008] Calculate the order to be allocated order priority index according to each order to be allocated, and obtain a set of scheduled cargo ships from the set of all cargo ships of the transporters; the order to be allocated order with the largest order to be allocated order priority index is the priority order to be allocated order, and the transport capacity is scheduled; the optimal scheduled cargo ship in the set of scheduled cargo ships is selected for the priority order to be allocated order, and a scheduled cargo ship order strategy is established;

[0009] If the optimal scheduled cargo ship coincides within the transport capacity scheduling relaxation time, execute a dynamic auction strategy to reschedule the optimal scheduled cargo ship, and reestablish the scheduled cargo ship order strategy.

[0010] Traverse all the to-be-allocated order set according to the size of the to-be-allocated order priority index, and complete intelligent transport capacity scheduling.

[0011] As a preferred technical solution of the present application, the specific steps of calculating the to-be-allocated order priority index according to each to-be-allocated order include:

[0012] Match the to-be-allocated order with a suitable cargo ship in the transport party's all cargo ship set to obtain a to-be-scheduled cargo ship set, and the to-be-scheduled cargo ship set contains M to-be-scheduled cargo ships;

[0013] Calculate the departure urgency factor based on the corresponding storage time limit and cargo attribute data of the transport port, calculate the transport complexity factor based on the cargo attribute data, and calculate the average transport green gain factor based on the real-time cabin capacity and green transport index of each to-be-scheduled cargo ship matched by the to-be-allocated order;

[0014] Weighted calculation of the departure urgency factor, the transport complexity factor and the average transport green gain factor to obtain the to-be-allocated order priority index of the to-be-allocated order.

[0015] As a preferred technical solution of the present application, the specific steps of selecting the optimal to-be-scheduled cargo ship in the to-be-scheduled cargo ship set for the priority to-be-allocated order include:

[0016] Based on the to-be-scheduled cargo ship set, M transport party game bodies T m , m=1, 2, …, M are constructed;

[0017] Define a transport capacity multi-objective evaluation index, which contains a first sub-target, a second sub-target, a third sub-target and a fourth sub-target; the first sub-target is the maximum cargo ship turnover rate; the second sub-target is the maximum cargo ship cabin utilization rate; the third sub-target is the maximum green transport capacity index; and the fourth sub-target is the maximum transport timeliness;

[0018] According to the group game rule and the transport capacity multi-objective evaluation index, evolutionary game is performed on all transport party game bodies T m , and the optimal transport party game body T m corresponding cargo ship is output as the optimal to-be-scheduled cargo ship, and a to-be-allocated order-cargo ship transport capacity scheduling strategy is established;

[0019] In each to-be-allocated order-cargo ship transport capacity scheduling strategy, there is a to-be-allocated order and a corresponding optimal to-be-scheduled cargo ship.

[0020] As a preferred technical solution of the present application, the specific steps of performing evolutionary game on all transport party game bodies T m according to the group game rule include:

[0021] Based on the route data of priority orders to be assigned, obtain all the transport ports along the way;

[0022] For the transportation party game entity T m Obtain the set B of alternative orders to be assigned from the ports of transit. m Output the transit port queue, where the transit port queue = {port 1, port 2, ..., port I}, and I represents the total number of ports in the transit port queue; based on the real-time water level data of the ports and the predicted water level values ​​in the next N time slices, select the set B of candidate orders to be assigned. m After re-filtering, a new set of candidate orders to be assigned, B, is obtained. m ';

[0023] Obtain the transport party's game entity T m The weighted value of the order grabbing Q m Based on the order-grabbing weighted value Q m A new set of alternative orders to be assigned, B m By combining swarm optimization algorithms to output multi-objective evaluation indicators for transport capacity, the optimal transport player game entity T is obtained. m .

[0024] As a preferred embodiment of the present invention, the specific steps for implementing a dynamic auction strategy to reschedule the optimal cargo ship to be scheduled include:

[0025] The scheduling orders corresponding to the overlapping optimal cargo ships to be scheduled are treated as independent intelligent agents Z. j j=1,2,…,J; J is the total number of orders to be scheduled corresponding to the overlap of the optimal cargo ships to be scheduled;

[0026] Set up an independent intelligent agent Z j Competition integral E j Each independent intelligent agent Z j A virtual bid is made for the optimal cargo ship to be scheduled, resulting in the agent's bid B. j ;

[0027] Obtaining the independent intelligent agent Z j The bidding weighted value R j According to the bidding weighted value R j Recalculation yields a new agent bid B. j ';

[0028] All new agent bidding B j The highest-ranking agent is selected as the optimal independent agent. Based on the optimal independent agent, the optimal cargo ship to be scheduled is matched, and the order-cargo ship capacity scheduling strategy is re-established.

[0029] As a preferred technical solution of the present invention, in evolutionary game theory, a swarm algorithm is combined to output multi-objective evaluation indicators of transportation capacity, and the swarm algorithm is a genetic algorithm.

[0030] An intelligent transportation capacity scheduling system includes:

[0031] The capacity information integration module includes an information extraction unit. This unit is used in the current capacity scheduling scenario, which includes several orders to be allocated, carriers, and transport ports. It acquires a set of orders to be allocated, extracting cargo attribute data, route data, and the corresponding storage time limit of each transport port for each order. It also acquires a set of all cargo ships belonging to each carrier, extracting the real-time cargo capacity, draft, and green transport index of each ship. Finally, it acquires a set of transport ports, extracting the real-time water level data and predicted water level values ​​for the next N time slices for each port.

[0032] The capacity scheduling module includes an order priority scheduling unit and a transportation adjustment scheduling unit. The order priority scheduling unit calculates the priority index of each pending order and obtains the set of cargo ships to be scheduled from all cargo ship sets of the transporter. The pending order with the highest priority index is designated as the priority pending order and is used for capacity scheduling. The optimal cargo ship to be scheduled is selected from the set of cargo ships to be scheduled for the priority pending orders, and a pending order-cargo ship capacity scheduling strategy is established. The transportation adjustment scheduling unit is used to execute a dynamic auction strategy to reschedule the optimal cargo ship to be scheduled if there is overlap of optimal cargo ships during the capacity scheduling buffer period, and to re-establish the pending order-cargo ship capacity scheduling strategy. It iterates through all pending order sets according to the priority index of the pending orders to complete intelligent capacity scheduling.

[0033] The present invention has the following advantages:

[0034] 1. This invention calculates the priority index of each order to be assigned, enabling the intelligent scheduling system to sort all orders according to their priority and identify the highest priority order. Priority sorting considers not only the urgency of the order but also comprehensively evaluates factors such as cargo attributes, cargo capacity, and green transportation, and forms a comprehensive priority index through weighted calculation, thereby improving the accuracy of scheduling decisions. Through group game rules, transporters optimize their bidding strategies based on order priority and scheduling needs, enhancing the flexibility and diversity of scheduling. During the game, transporters adjust their bidding weights according to different strategies, allowing all parties to dynamically adjust scheduling plans under limited resources.

[0035] 2. This invention, by acquiring and predicting port water level data in real time, can adjust scheduling strategies based on future water level changes, avoiding transportation delays caused by water level issues and enhancing the system's adaptability to dynamic environmental changes. When the optimal cargo ship overlaps with multiple orders, the system adopts a dynamic auction strategy. The scheduling scheme is readjusted through a bidding mechanism to ensure that high-priority orders are scheduled first, while avoiding resource conflicts and optimizing capacity allocation. The auction strategy is optimized based on the bid and priority of each order, ensuring optimal utilization of limited resources. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the structure of an intelligent transportation capacity scheduling system used in an embodiment of the present invention. Detailed Implementation

[0037] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this invention.

[0038] Example 1: An intelligent transportation capacity scheduling method, comprising the following steps:

[0039] In the current capacity scheduling scenario, there are several orders to be allocated, transporters, and transport ports;

[0040] Get the set of orders to be assigned, and extract the cargo attribute data, route data and corresponding storage time limit of the port of transport for each order; get the set of all cargo ships of the transporter, and extract the real-time capacity, draft and green transport index of each cargo ship; get the set of transport ports, and extract the real-time water level data and the predicted water level value in the next N time slices for each transport port.

[0041] Cargo attribute data provides a detailed description of the physical characteristics, category, and special transportation requirements of the cargo to be transported, including cargo type (such as general cargo, refrigerated cargo, dangerous goods, etc.), weight (unit: tons / t), volume (unit: cubic meters / m³), packaging form (such as container, bulk), temperature control requirements (such as -20℃ to 5℃), shock resistance level, and chemical properties (such as flammability, corrosiveness); it is used to determine whether the cargo ship has the corresponding hold space and to calculate the cargo ship's load balance based on weight and volume to prevent uneven loading;

[0042] Route data defines the transportation route of goods from the port of origin to the port of destination, including geographical coordinates, ports of transit, distance of the route, and estimated sailing time. It is used for cost estimation, transportation risk assessment, and dynamic adjustment of transportation.

[0043] Storage time limit refers to the maximum storage time that goods are allowed in the port. Exceeding the time limit will result in demurrage fees or forced transshipment. It is used to prioritize the allocation of fast cargo ships for orders with close time limits, avoid high demurrage fees, provide early warning of port storage pressure, and coordinate loading and unloading resources.

[0044] Real-time cargo capacity indicates the real-time data of the cargo ship's currently available loading space, measured in deadweight tons or cubic meters; cargo ship draft indicates the vertical depth of the cargo ship when it is stationary and submerged in water, including empty draft and full load draft. For example, a cargo ship has an empty draft of 6.5 meters and a full load draft of 12 meters. This is used to match the real-time water level of the port and, together with the load line markings, to calculate the maximum cargo capacity. During low tide, only cargo ships with shallower drafts are allowed to enter the port.

[0045] Real-time water level data represents the current actual water depth of the transport port; the water level prediction value in the next N time slices is based on the current astronomical tides, meteorological data and river runoff to predict future water level changes, and the size of the time slice can be set to 1 hour;

[0046] The formula for calculating the Green Transport Index is L = (ship turnover rate / benchmark value) * (1 - carbon emission intensity / average carbon emission) * fuel efficiency rating;

[0047] Cargo ship turnover rate indicates the number of times a cargo ship completes transportation tasks per unit of time (such as monthly / quarterly); for example, if a cargo ship completes 5 "originating port to destination port" transportations per month, then the turnover rate is 5 times / month; the benchmark value is the average turnover rate of similar cargo ship types, which is set manually by professional technicians.

[0048] Carbon emission intensity represents the amount of carbon dioxide emitted per unit of transport volume, with the unit of transport volume being tons per nautical mile; carbon dioxide emissions are calculated from fuel consumption and fuel carbon conversion factor; the average carbon emission is the average carbon emission of similar cargo ship types, which is set manually by professional technicians.

[0049] Fuel efficiency rating indicates a rating of a cargo ship’s fuel consumption efficiency per unit of transport volume. Under industry standard rules, the rating is based on the cargo ship’s historical fuel consumption data. The higher the green transport index, the better the cargo ship’s overall performance in terms of efficiency, environmental protection, and cost.

[0050] Calculate the priority index of each pending order, and obtain the set of cargo ships to be dispatched from all cargo ship sets of the transporter; the pending order with the highest priority index is designated as the priority pending order and is used for capacity dispatch; select the optimal cargo ship to be dispatched from the set of cargo ships to be dispatched for the priority pending order, and establish a pending order-cargo ship capacity dispatch strategy.

[0051] The specific steps for calculating the priority index for each pending order include:

[0052] For each order to be assigned, a matching cargo ship is found in the carrier's set of all cargo ships to obtain a set of cargo ships to be dispatched, which contains M cargo ships to be dispatched.

[0053] The departure urgency factor is calculated based on the corresponding storage time limit and cargo attribute data of the port of origin; the transportation complexity factor is calculated based on the cargo attribute data; and the average transportation green gain factor is calculated based on the real-time capacity of each cargo ship to be dispatched and the green transportation index matched with the orders to be assigned.

[0054] The weighted calculation of departure urgency factor, transportation complexity factor and average transportation green gain factor yields the order priority index for orders to be allocated.

[0055] The specific steps for calculating the departure urgency factor are as follows: The formula for calculating the departure urgency factor is: (1 - Remaining storage time / Total storage time) * (1 + α * Cargo Specificity Coefficient); where the remaining storage time is determined by the corresponding storage time of the port of transport to which the order to be allocated belongs, and the total storage time is determined by the special transportation requirements in the cargo attribute data; the cargo specificity coefficient is allocated based on the cargo type, for example, dangerous goods = 0.5, refrigerated goods = 0.3, general goods = 0, and the corresponding rules are set manually; α represents the urgency adjustment factor, which is determined manually;

[0056] The specific steps for calculating the transportation complexity factor are as follows: The formula for calculating the transportation complexity factor is weight / baseline weight, volume / baseline volume, and cargo type complexity, which are weighted and calculated. The baseline weight and baseline volume are set manually. The cargo type complexity is allocated based on the cargo type, for example, dangerous goods = 1.2, refrigerated goods = 1.0, and general goods = 0.8, with corresponding rules set manually.

[0057] The specific steps for calculating the average green transport gain factor are as follows: Calculate the green transport index corresponding to each cargo ship to be dispatched that matches the orders to be assigned.

[0058] The core function of the departure urgency factor is to ensure the most efficient completion of time-sensitive orders, especially those requiring priority processing due to approaching storage deadlines, urgent needs, or special cargo types (such as dangerous goods or refrigerated cargo). This factor helps prioritize the scheduling of goods that are about to expire or have high timeliness requirements, avoiding storage delays, cargo losses, or additional costs. The transport complexity factor is mainly used to assess the transport difficulty of cargo and the required resource input. More complex cargoes (such as heavy, large, or special transport requirements) usually require larger cargo space, more efficient transport routes, or more complex logistics support. Therefore, this factor helps determine which cargo ships can carry these complex cargoes, avoiding unsuitable cargo ships carrying highly complex orders. The transport green gain factor is used to measure the environmental performance of each cargo ship to be scheduled, especially its environmental impact (such as carbon emissions, energy efficiency, etc.). By calculating the green gain factor, it is possible to prioritize the selection of environmentally friendly vessels during capacity scheduling, optimizing the environmental benefits of the entire transport process.

[0059] The specific steps for selecting the optimal cargo ship to be scheduled from the set of priority pending orders include:

[0060] Based on the set of cargo ships to be scheduled, construct a game of M transportation parties, T. m m = 1, 2, ..., M;

[0061] A multi-objective evaluation index for capacity is defined, which includes a first sub-objective, a second sub-objective, a third sub-objective, and a fourth sub-objective; the first sub-objective is the maximum cargo ship turnover rate; the second sub-objective is the maximum cargo ship capacity utilization rate; the third sub-objective is the maximum green capacity index; and the fourth sub-objective is the maximum transportation timeliness.

[0062] Based on the group game rules and multi-objective evaluation indicators of transport capacity, T is the game entity for all transport parties. m Perform evolutionary game theory to output the optimal transporter game state T. m The corresponding cargo ship is selected as the optimal cargo ship to be dispatched, and a scheduling strategy for pending orders and cargo ship capacity is established.

[0063] Each pending order-shipping capacity scheduling strategy corresponds to one pending order and one optimal cargo ship to be scheduled.

[0064] According to the rules of group game theory, the game entities T for all transport parties are... m The specific steps for conducting evolutionary game theory include:

[0065] Based on the route data of priority orders to be assigned, obtain all the transport ports along the way;

[0066] For each order to be assigned, extract its origin port and destination port of transport, as well as multiple ports that may be passed through along the way. In the route data, each route may contain multiple transit ports. Extract all ports that are passed through from the route data. Routes usually contain multiple points, so it is necessary to identify the ports that each order passes through and build a port list.

[0067] For the transportation party game entity T m Obtain the set B of alternative orders to be assigned from the ports of transit. m Output the transit port queue, where the transit port queue = {port 1, port 2, ..., port I}, and I represents the total number of ports in the transit port queue; based on the real-time water level data of the ports and the predicted water level values ​​in the next N time slices, select the set B of candidate orders to be assigned. m After re-filtering, a new set of candidate orders to be assigned, B, is obtained. m ';

[0068] Extract all pending orders whose originating port of transport is a transit port, and whose cargo attribute data satisfies the transport party's game theory T. m The remaining cargo space, for each transport party's game entity T m Based on the current available space and the demand for orders, suitable alternative orders need to be selected for allocation. By checking the port of origin, port of destination, and possible transit ports for each order, it is determined whether each order can be transported through the current transit ports. Combining the remaining space data, those orders that meet the transportation requirements and the current remaining space on the cargo ship are selected to obtain the set of alternative orders for allocation.

[0069] For each candidate order to be assigned, the real-time water level and the water level forecast for the next N time slices are checked to see if they meet the requirements based on the water level requirements of the port of arrival. Only if the water level forecast meets the requirements will the order be selected to enter the re-filtered set of candidate orders to be assigned.

[0070] Obtain the transport party's game entity T m The weighted value of the order grabbing Q m Based on the order-grabbing weighted value Q m A new set of alternative orders to be assigned, B m By combining swarm optimization algorithms to output multi-objective evaluation indicators for transport capacity, the optimal transport player game entity T is obtained. m ;

[0071] The order-grabbing weighting value is operated by the transportation party. The order-grabbing operation is carried out according to the transportation party's order-grabbing operation. For example, the order-grabbing operation can be carried out by reducing the single transportation cost, and the order-grabbing weighting value is assigned proportionally.

[0072] Based on the group algorithm, the game and competition between transport parties are simulated. The performance of each transport party is evaluated using the multi-objective evaluation index of transport capacity to obtain the corresponding fitness. Based on the corresponding fitness, the optimal transport party game entity is selected, that is, the best performing cargo ship and transport strategy are selected.

[0073] In evolutionary game theory, a swarm algorithm is used to output multi-objective evaluation indicators of transportation capacity. The swarm algorithm is a genetic algorithm.

[0074] In the capacity scheduling game, multiple transport parties compete according to game rules to achieve the optimal cargo ship scheduling scheme. First, relevant parameters are initialized for each transport party, including the set of orders to be allocated, remaining space information, and port data along the route. Next, based on the route data of each order, all transport ports along the route are extracted, and real-time water level data and water level predictions for the next N time slices are obtained for each port. This is used to determine which orders can pass the water level restrictions of specific ports, and a new set of candidate orders that meet the conditions is selected. Each transport party performs multi-objective optimization based on its order-grabbing weighting value, combined with a swarm optimization algorithm, such as genetic algorithm or particle swarm optimization. The objective functions include maximizing cargo ship turnover rate, cargo ship space utilization rate, green capacity index, and transportation timeliness. The comprehensive score of each transport party is calculated through multi-objective evaluation indicators of capacity. Based on these evaluation indicators, transport companies select the best routes and cargo ship resources and participate in a game. During the game, transport companies compete with each other, simulating a bidding mechanism and adjusting their strategies to obtain the optimal scheduling plan. After each round of the game, transport companies update their decisions based on the game results and feedback on strategy adjustments to adapt to changes in the market and actual demand. Finally, after multiple rounds of the game and strategy optimization, all transport companies will output an optimal scheduling strategy, and the cargo ship corresponding to the optimal transport company is the optimal cargo ship to be scheduled.

[0075] If the optimal cargo ships to be dispatched overlap during the period of relaxed capacity scheduling, the dynamic auction strategy will be implemented to reschedule the optimal cargo ships to be dispatched and a new order-cargo ship capacity scheduling strategy will be established.

[0076] The buffer time for capacity scheduling is set manually;

[0077] The system iterates through all sets of orders to be assigned according to their priority index to complete intelligent capacity scheduling.

[0078] The specific steps for implementing a dynamic auction strategy to reschedule the optimal cargo ship include:

[0079] The scheduling orders corresponding to the overlapping optimal cargo ships to be scheduled are treated as independent intelligent agents Z. j j=1,2,…,J; J is the total number of orders to be scheduled corresponding to the overlap of the optimal cargo ships to be scheduled;

[0080] Set up an independent intelligent agent Z j Competition integral E j Each independent intelligent agent Z j A virtual bid is made for the optimal cargo ship to be scheduled, resulting in the agent's bid B. j The calculation steps for the competition score are as follows: the result of time urgency factor * transportation difficulty factor * ln(1 + green adaptation factor) corresponding to the order to be scheduled is used as the initial competition score; calculate the route overlap degree between each independent agent and the optimal cargo ship to be scheduled; multiply the initial competition score by the route overlap degree to obtain the agent bidding.

[0081] The route overlap calculation steps are as follows: First, extract the route information of the order and the historical route data of the candidate cargo ships from the data. The matching degree between each route and the current order is evaluated by comparing the similarity of the origin and destination, ports passed through, and distance of the historical route data. Specifically, the matching of origin and destination, the number of intersections of ports passed through, and the difference in route distance are calculated to obtain a base score. The base score is adjusted in combination with the cargo ship's historical performance (such as the number of executions and success rate) to obtain the final overlap degree. The higher the final overlap degree, the higher the matching degree between the cargo ship and the order, and the better the suitability.

[0082] Obtaining the independent intelligent agent Z j The bidding weighted value R j According to the bidding weighted value R j Recalculation yields a new agent bid B. j This is equivalent to each order to be assigned bidding for a cargo ship according to its priority and bidding rules. The bid reflects the value of the order obtaining cargo ship scheduling. The bidding weighting value is the increase in the single transportation price that each independent agent can accept, and the bidding weighting value is assigned proportionally.

[0083] All new agent bidding B j The highest-ranking agent is selected as the optimal independent agent. Based on the optimal independent agent, the optimal cargo ship to be scheduled is matched, and the order-cargo ship capacity scheduling strategy is re-established.

[0084] The overlap of optimal scheduled cargo ships typically occurs when cargo ship resources are limited during transportation and multiple scheduling tasks need to be completed simultaneously. In actual capacity scheduling, the speed at which cargo ships load cargo is not always synchronized with the speed of scheduling calculations. Scheduling calculations are usually based on ideal speeds to deduce the optimal scheduling scheme, but in actual operation, loading speed is often affected by various factors such as port operations, cargo type, and cargo ship facilities. Therefore, the calculation speed may be too fast, causing the scheduling time of some cargo ships to be advanced, which may result in the same cargo ship being assigned multiple tasks during the scheduling process. In actual transportation, the scheduling between cargo ships may not be completely tight, as operations such as cargo ships arriving at the port, unloading, and reloading require... For a certain period of time, this provides space for overlap of cargo ship resources during the scheduling process. During this buffer period, the transporter may plan to adjust the task order of cargo ships or temporarily allocate other tasks to ensure that each task can be carried out smoothly. This results in a cargo ship "overlapping" with multiple pending orders in time. After determining the optimal independent agent and matching the optimal cargo ship through the bidding process, the capacity scheduling strategy needs to be updated. The matching relationship between each pending order and cargo ship should be re-planned, taking into account the priority of each order and the transport capacity of the cargo ship. Through the bidding mechanism and weighted rules in the game, it is ensured that only the most competitive order can obtain cargo ship scheduling, and the situation of multiple orders competing for the same cargo ship is avoided. This can effectively solve resource conflicts and optimize the task allocation of each cargo ship.

[0085] Example 2, an intelligent transportation capacity scheduling system, see [link to example]. Figure 1 As shown, it includes:

[0086] The capacity information integration module includes an information extraction unit. This unit is used in the current capacity scheduling scenario, which includes several orders to be allocated, carriers, and transport ports. It acquires a set of orders to be allocated, extracting cargo attribute data, route data, and the corresponding storage time limit of each transport port for each order. It also acquires a set of all cargo ships belonging to each carrier, extracting the real-time cargo capacity, draft, and green transport index of each ship. Finally, it acquires a set of transport ports, extracting the real-time water level data and predicted water level values ​​for the next N time slices for each port.

[0087] The capacity scheduling module includes an order priority scheduling unit and a transportation adjustment scheduling unit. The order priority scheduling unit calculates the priority index of each pending order and obtains the set of cargo ships to be scheduled from all cargo ship sets of the transporter. The pending order with the highest priority index is designated as the priority pending order and is used for capacity scheduling. The optimal cargo ship to be scheduled is selected from the set of cargo ships to be scheduled for the priority pending orders, and a pending order-cargo ship capacity scheduling strategy is established. The transportation adjustment scheduling unit is used to execute a dynamic auction strategy to reschedule the optimal cargo ship to be scheduled if there is overlap of optimal cargo ships during the capacity scheduling buffer period, and to re-establish the pending order-cargo ship capacity scheduling strategy. It iterates through all pending order sets according to the priority index of the pending orders to complete intelligent capacity scheduling.

[0088] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims. Parts not described in detail in this specification are prior art known to those skilled in the art.

Claims

1. An intelligent transportation capacity scheduling method, characterized in that, Includes the following steps: In the current capacity scheduling scenario, there are several orders to be allocated, transporters, and transport ports; Obtain the set of orders to be assigned, and extract the cargo attribute data, route data, and corresponding storage time limit of the port of transport for each order to be assigned; Get the set of all cargo ships of the transporter, and extract the real-time cargo capacity, draft and green transport index of each cargo ship; get the set of transport ports, and extract the real-time water level data and the water level prediction value in the next N time slices for each transport port. Calculate the priority index of each pending order based on the pending order, and obtain the set of cargo ships to be dispatched from all cargo ship sets of the transporter; The departure urgency factor is calculated based on the corresponding storage time limit and cargo attribute data of the port of origin. Calculate transportation complexity factors based on cargo attribute data; The average transportation green gain factor is calculated based on the real-time capacity of each cargo ship to be dispatched and the green transportation index matched with the orders to be assigned. The order with the highest priority index is designated as the priority order and is used for capacity scheduling. The optimal cargo ship in the set of cargo ships to be scheduled is selected for the priority order, and a capacity scheduling strategy for the order-cargo ship is established. If the optimal cargo ships to be dispatched overlap during the period of relaxed capacity scheduling, the dynamic auction strategy will be implemented to reschedule the optimal cargo ships to be dispatched and a new order-ship capacity scheduling strategy will be established. The system iterates through all sets of orders to be assigned according to their priority index to complete intelligent capacity scheduling. The specific steps for selecting the optimal cargo ship to be scheduled from the set of priority pending orders include: Based on the set of cargo ships to be scheduled, construct a game of M transportation parties, T. m m = 1, 2, ..., M; A multi-objective evaluation index for capacity is defined, which includes a first sub-objective, a second sub-objective, a third sub-objective, and a fourth sub-objective; the first sub-objective is the maximum cargo ship turnover rate; the second sub-objective is the maximum cargo ship capacity utilization rate; the third sub-objective is the maximum green capacity index; and the fourth sub-objective is the maximum transportation timeliness. Based on the group game rules and multi-objective evaluation indicators of transport capacity, T is the game entity for all transport parties. m Perform evolutionary game theory to output the optimal transporter game state T. m The corresponding cargo ship is selected as the optimal cargo ship to be dispatched, and a scheduling strategy for pending orders and cargo ship capacity is established. Each pending order-shipping capacity scheduling strategy corresponds to one pending order and one optimal cargo ship to be scheduled. The specific steps for implementing a dynamic auction strategy to reschedule the optimal cargo ship include: The scheduling orders corresponding to the overlapping optimal cargo ships to be scheduled are treated as independent intelligent agents Z. j j=1,2,…,J; J is the total number of orders to be scheduled corresponding to the overlap of the optimal cargo ships to be scheduled; Set up an independent intelligent agent Z j Competition integral E j Each independent intelligent agent Z j A virtual bid is made for the optimal cargo ship to be scheduled, resulting in the agent's bid B. j ; The steps for calculating the competition score are as follows: the result of time urgency factor * transportation difficulty factor * ln(1 + green adaptation factor) corresponding to the order to be scheduled is used as the initial competition score; calculate the route overlap degree between each independent agent and the optimal cargo ship to be scheduled; multiply the initial competition score by the route overlap degree to obtain the agent's bid. The steps for calculating route overlap are as follows: First, extract the route information of the order and the historical route data of the candidate cargo ships from the data. The matching degree between each route and the current order is evaluated by comparing the similarity of the origin and destination, ports passed through, and distances of the historical route data. Specifically, the matching of origin and destination, the number of intersections of ports passed through, and the difference in route distance are calculated to obtain a base score. The base score is then adjusted based on the historical performance of the cargo ships to obtain the final overlap. Obtaining the independent intelligent agent Z j The bidding weighted value R j According to the bidding weighted value R j Recalculation yields a new agent bid B. j '; All new agent bidding B j The highest-ranking agent is selected as the optimal independent agent. Based on the optimal independent agent, the optimal cargo ship to be scheduled is matched, and the order-cargo ship capacity scheduling strategy is re-established.

2. The intelligent transportation capacity scheduling method according to claim 1, characterized in that, The specific steps for calculating the priority index for each pending order include: For each order to be assigned, a matching cargo ship is found in the carrier's set of all cargo ships to obtain a set of cargo ships to be dispatched, which contains M cargo ships to be dispatched. The weighted calculation of departure urgency factor, transportation complexity factor and average transportation green gain factor yields the priority index of pending orders.

3. The intelligent transportation capacity scheduling method according to claim 2, characterized in that, According to the rules of group game theory, the game entities T for all transport parties are... m The specific steps for conducting evolutionary game theory include: Based on the route data of priority orders to be assigned, obtain all the transport ports along the way; For the transportation party game entity T m Obtain the set B of alternative orders to be assigned from the ports of transit. m Output the transit port queue, where the transit port queue = {port 1, port 2, ..., port I}, and I represents the total number of ports in the transit port queue; based on the real-time water level data of the ports and the predicted water level values ​​in the next N time slices, select the set B of candidate orders to be assigned. m After re-filtering, a new set of candidate orders to be assigned, B, is obtained. m '; Obtain the transport party's game entity T m The weighted value of the order grabbing Q m Based on the order-grabbing weighted value Q m A new set of alternative orders to be assigned, B m By combining swarm optimization algorithms to output multi-objective evaluation indicators for transport capacity, the optimal transport player game entity T is obtained. m .

4. The intelligent transportation capacity scheduling method according to claim 3, characterized in that, In evolutionary game theory, a swarm algorithm is used to output multi-objective evaluation indicators for transportation capacity. The swarm algorithm is a genetic algorithm.

5. An intelligent transportation capacity scheduling system, characterized in that, The system employs an intelligent transportation capacity scheduling method according to any one of claims 1-4, comprising: The capacity information integration module includes an information extraction unit. This unit is used in the current capacity scheduling scenario, which includes several orders to be allocated, carriers, and transport ports. It acquires a set of orders to be allocated, extracting cargo attribute data, route data, and the corresponding storage time limit of each transport port for each order. It also acquires a set of all cargo ships belonging to each carrier, extracting the real-time cargo capacity, draft, and green transport index of each ship. Finally, it acquires a set of transport ports, extracting the real-time water level data and predicted water level values ​​for the next N time slices for each port. The capacity scheduling module includes an order priority scheduling unit and a transportation adjustment scheduling unit. The order priority scheduling unit calculates the priority index of each pending order and obtains the set of cargo ships to be scheduled from all cargo ship sets of the transporter. The pending order with the highest priority index is designated as the priority pending order and is used for capacity scheduling. The optimal cargo ship to be scheduled is selected from the set of cargo ships to be scheduled for the priority pending orders, and a pending order-cargo ship capacity scheduling strategy is established. The transportation adjustment scheduling unit is used to execute a dynamic auction strategy to reschedule the optimal cargo ship to be scheduled if there is overlap of optimal cargo ships during the capacity scheduling buffer period, and to re-establish the pending order-cargo ship capacity scheduling strategy. It iterates through all pending order sets according to the priority index of the pending orders to complete intelligent capacity scheduling.

Images